The Prediction of Free Radical Scavenging Activity, Starter Cultures Count and Sensory Characteristics of Probiotic Yogurt Containing the Hydroalcoholic Extracts of Spirulinaplatensis and Ferulagoangulata by Artificial Neural Network (ANN)

The purpose of this study was to evaluate the free radical scavenging activity, yogurt starters counts and sensory characteristics (color, flavour, consistency, overall acceptance) of low-fat probiotic yogurt enriched with different levels of hydroalcoholic extract of Ferulago angulata (F.angulata) and microalgae Spirulina platensis (S. platensis) for 21 days at refrigerated temperature and predicting the experiments results by Artificial neural network (ANN). In order to predict the experiments results, the multilayer perceptron neural network with two inputs (extract concentration, storage time) and one output using MATLAB R2013a software was used. About free radical scavenging activity (DPPH) of probiotic yogurt samples, the R2 was determined to be 0/989. Over time, by the 14th day, the free radical scavenging activity was increased and then decreased. F. angulata extract had a greater impact on this activity. For L. bulgaricus and S. thermophilus, R2 were 0/99 and 0/97 respectively. Treatments containing S. platensis extract had a higher count of these strains in comparsion with F. angulata extract. while the starter cultures count in the final day was determined to be more than the minimum number in probiotic yogurt (107 cfu/ml). R2 for color, flavor, consistency, and overall acceptance were 0/93, 0/98, 0/98, and 0/97 respectively, while the treatments containing S. platensis extract showed higher sensory scores in comparsion with F. angulata extract. The results showed that using a network with 10 neurons in the hidden layer and by the active hyperbolic sigmoid function and the data percentage used for training / test / evaluation is 60/ 15 / 20, probiotic yogurt can be predicted. The sensitivity analysis of optimal neural network showed the importance of predicting the parameters of extract concentrations and storage time on the changes of physicochemical, microbial and sensory characteristics of probiotic yogurt. It can be said ANN is a powerful tool for predicting the quality of probiotic yogurt.